Svetha Venkatesh

Deakin University · Institute for Frontier Materials (IFM)

At the heart of the flourishing field of machine learning potentials are graph neural networks, where deep learning is interwoven with physics-informed machine learning (PIML) architectures. Various PIML models, upon training with density functional theory (DFT) material structure–property datasets, have achieved unprecedented prediction accuracy f...

Discovering new solid-state materials requires rapidly exploring the vast space of crystal structures and locating stable regions. Generating stable materials with desired properties and compositions is extremely difficult as we search for very small isolated pockets in the exponentially many possibilities, considering elements from the periodic ta...

Although lithium solid state electrolytes promise to mitigate the chemical instabilities of liquid electrolytes in today's mainstream rechargeable batteries, solid state electrolytes still suffer from dendrite formation which leads to battery degradation and short circuiting. Dendrite initiation and propagation in specific solid state electrolyte m...

In this paper, we study the “dataset bias” problem from a statistical standpoint, and identify the main cause of the problem as the strong correlation between a class attribute u and a non-class attribute b in the input x, represented by p(u|b) differing significantly from p(u). Since p(u|b) appears as part of the sampling distributions in the stan...

Current lithium batteries do not fully meet the longevity and safety requirements of electric vehicles. Novel solid state lithium ion batteries could be a compelling solution to these problems. In this work we unravel some of these new materials with potentially high lithium conductivity by using a Bayesian optimization approach. This involves expl...

Prompt optimization is essential for enhancing the performance of Large Language Models (LLMs) in a range of Natural Language Processing (NLP) tasks, particularly in scenarios of few-shot learning where training examples are incorporated directly into the prompt. Despite the growing interest in optimizing prompts with few-shot examples, existing me...

Effective decision-making in partially observable environments demands robust memory management. Despite their success in supervised learning, current deep-learning memory models struggle in reinforcement learning environments that are partially observable and long-term. They fail to efficiently capture relevant past information, adapt flexibly to...

Typically developing infants, between the corrected age of 9–20 weeks, produce fidgety movements. These movements can be identified with the General Movement Assessment, but their identification requires trained professionals to conduct the assessment from video recordings. Since trained professionals are expensive and their demand may be higher th...

3D bioprinting enables the fabrication of printable tissues, including those for neural, cartilage, and skin repair. The mechanical properties, especially stiffness, of 3D-bioprinted scaffolds are crucial for modulating cell adhesion, growth, migration, and differentiation. The stiffness of a scaffold can be adjusted post-printing by modifying the...

Introduction University students are particularly susceptible to developing high levels of stress, which occur when environmental demands outweigh an individual’s ability to cope. The growing advent of mental health smartphone apps has led to a surge in use by university students seeking ways to help them cope with stress. Use of these apps has aff...

Prompt optimization is essential for enhancing the performance of Large Language Models (LLMs) in a range of Natural Language Processing (NLP) tasks, particularly in scenarios of few-shot learning where training examples are incorporated directly into the prompt. Despite the growing interest in optimizing prompts with few-shot examples, existing me...

The challenge in constructing artificial social agents is to enable adaptation ability to novel agents, and is called zero-shot coordination (ZSC). A promising approach is to train the adaptive agents by interacting with a diverse pool of collaborators, assuming that the greater the diversity in other agents seen during training, the better the gen...

Modern reinforcement learning (RL) struggles to capture real-world cause-and-effect dynamics, leading to inefficient exploration due to extensive trial-and-error actions. While recent efforts to improve agent exploration have leveraged causal discovery, they often make unrealistic assumptions of causal variables in the environments. In this paper,...

Background More than 50% of people who die by suicide have not been in contact with formal mental health services. The rate of people who fly ‘under the radar’ of mental health services is higher among men than women, indicating a need to improve engagement strategies targeted towards men who experience suicidal thoughts and/or behaviours. In Austr...

Learning composite concepts, such as \textquotedbl red car\textquotedbl , from individual examples -- like a white car representing the concept of \textquotedbl car\textquotedbl{} and a red strawberry representing the concept of \textquotedbl red\textquotedbl -- is inherently challenging. This paper introduces a novel method called Composite Concep...

BACKGROUND Prevention of suicide is a global health priority. Around 800,000 individuals die by suicide yearly, and for every death, there are another 20 estimated suicide attempts. Large language models (LLMs) hold the potential to enhance scalable, accessible, and affordable digital services for suicide prevention and self-harm interventions. How...

How can Large Language Models (LLMs) be aligned with human intentions and values? A typical solution is to gather human preference on model outputs and finetune the LLMs accordingly while ensuring that updates do not deviate too far from a reference model. Recent approaches, such as direct preference optimization (DPO), have eliminated the need for...

This paper presents two models of neural-networks and their training applicable to neural networks of arbitrary width, depth and topology, assuming only finite-energy neural activations; and a novel representor theory for neural networks in terms of a matrix-valued kernel. The first model is exact (un-approximated) and global, casting the neural ne...

BACKGROUND With increasing adoption of remote clinical trials in digital mental health, identifying recruitment methodologies that are both cost-effective and time-efficient is crucial for the success of such trials. Past research suggests that overall, there is no consistent evidence on whether web-based recruitment methods are more effective than...

Identifying root causes of anomalies in causal processes is vital across disciplines. Once identified, one can isolate the root causes and implement necessary measures to restore the normal operation. Causal processes are often modelled as graphs with entities being nodes and their paths/interconnections as edge. Existing work only consider the con...

Multivariate time series forecasting is essential in various fields, including healthcare and traffic management, but it is a challenging task due to the strong dynamics in both intra-channel relations (temporal patterns within individual variables) and inter-channel relations (the relationships between variables), which can evolve over time with a...

Using geotagged Twitter data in Victoria, we created a mobility index and studied the changes during the staged restrictions during the coronavirus disease 2019 (COVID-19) pandemic. We describe preliminary evidence that geotagged Twitter data may be used to provide real-time population mobility data and information on the impact of restrictions on...

Background Advances in smartphone technology have allowed people to access mental healthcare via digital apps from wherever and whenever they choose. University students experience a high burden of mental health concerns. Although these apps improve mental health symptoms, user engagement has remained low. Studies have shown that users can be subgr...

Large language models (LLMs) have recently demonstrated their impressive ability to provide context-aware responses via text. This ability could potentially be used to predict plausible solutions in sequential decision making tasks pertaining to pattern completion. For example, by observing a partial stack of cubes, LLMs can predict the correct seq...

We present a new computing model for intrinsic rewards in reinforcement learning that addresses the limitations of existing surprise-driven explorations. The reward is the novelty of the surprise rather than the surprise norm. We estimate the surprise novelty as retrieval errors of a memory network wherein the memory stores and reconstructs surpris...

University students are particularly susceptible to developing high levels of stress, which occur when environmental demands outweigh an individual's ability to cope. The growing advent of mental health smartphone apps has led to a surge in use by university students seeking ways to help them cope with stress. Use of these apps has afforded researc...

Physics-based reinforcement learning tasks can benefit from simplified physics simulators as they potentially allow near-optimal policies to be learned in simulation. However, such simulators require the latent factors (e.g. mass, friction coefficient etc.,) of the associated objects and other environment-specific factors (e.g. wind speed, air dens...

Model selection is an integral problem of model based optimization techniques such as Bayesian optimization (BO). Current approaches often treat model selection as an estimation problem, to be periodically updated with observations coming from the optimization iterations. In this paper, we propose an alternative way to achieve both efficiently. Spe...

Understanding other agents is a key challenge in constructing artificial social agents. Current works focus on centralised training, wherein agents are allowed to know all the information about others and the environmental state during training. In contrast, this work studies decentralised training, wherein agents must learn the model of other agen...

Fidgety movements occur in infants between the age of 9 to 20 weeks post-term, and their absence are a strong indicator that an infant has cerebral palsy. Prechtl's General Movement Assessment method evaluates whether an infant has fidgety movements, but requires a trained expert to conduct it. Timely evaluation facilitates early interventions, and...

Humans are highly adaptable, swiftly switching between different modes to progressively handle different tasks, situations and contexts. In Human-object interaction (HOI) activities, these modes can be attributed to two mechanisms: (1) the large-scale consistent plan for the whole activity and (2) the small-scale children interactive actions that s...

Introduction Meta-analytical evidence confirms a range of interventions, including mindfulness, physical activity and sleep hygiene, can reduce psychological distress in university students. However, it is unclear which intervention is most effective. Artificial intelligence (AI)-driven adaptive trials may be an efficient method to determine what w...

Simulation based learning often provides a cost-efficient recourse to reinforcement learning applications in robotics. However, simulators are generally incapable of accurately replicating real-world dynamics, and thus bridging the sim2real gap is an important problem in simulation based learning. Current solutions to bridge the sim2real gap involv...

The study of Neural Tangent Kernels (NTKs) has provided much needed insight into convergence and generalization properties of neural networks in the over-parametrized (wide) limit by approximating the network using a first-order Taylor expansion with respect to its weights in the neighborhood of their initialization values. This allows neural netwo...

Social reasoning necessitates the capacity of theory of mind (ToM), the ability to contextualise and attribute mental states to others without having access to their internal cognitive structure. Recent machine learning approaches to ToM have demonstrated that we can train the observer to read the past and present behaviours of other agents and inf...

The identification of the ground state phases of a chemical space in the convex hull analysis is a key determinant of the synthesizability of materials. Online material databases have been instrumental in exploring one aspect of the synthesizability of many materials, namely thermodynamic stability. However, the vibrational stability, which is anot...

Advances in smartphone technology have allowed people to access mental healthcare via digital apps from wherever and whenever they choose. University students experience a high burden of mental health concerns. Although these apps improve mental health symptoms, user engagement has remained low. Studies have shown that users can be subgrouped based...

Introduction: Meta-analytic evidence confirms a range of in- terventions, including mindfulness, physical activity and sleep hygiene, can reduce psychological distress in university stu- dents. However, it is unclear which intervention is most ef- fective. Artificial intelligence (AI) driven adaptive trials may be an efficient method to determine w...

Objectives The Future Proofing Study (FPS) was established to examine factors associated with the onset and course of mental health conditions during adolescence. This paper describes the design, methods, and baseline characteristics of the FPS cohort. Methods The FPS is an Australian school‐based prospective cohort study with an embedded cluster‐...

At the 2012 Hypervelocity Impact Symposium we demonstrated the utility of artificial neural networks (ANNs) for predicting the outcome of micrometeoroid and orbital debris (MMOD) impact on a Whipple shield at hypervelocity [1]. We established that machine learning (ML) techniques like ANN are well suited to high dimensionality problems like MMOD im...

Sample-efficient offline reinforcement learning (RL) with linear function approximation has recently been studied extensively. Much of prior work has yielded the minimax-optimal bound of $\tilde{\mathcal{O}}(\frac{1}{\sqrt{K}})$, with $K$ being the number of episodes in the offline data. In this work, we seek to understand instance-dependent bounds...

Trojan attacks on deep neural networks are both dangerous and surreptitious. Over the past few years, Trojan attacks have advanced from using only a single input-agnostic trigger and targeting only one class to using multiple, input-specific triggers and targeting multiple classes. However, Trojan defenses have not caught up with this development....

Knowledge distillation (KD) is an efficient approach to transfer the knowledge from a large “teacher” network to a smaller “student” network. Traditional KD methods require lots of labeled training samples and a white-box teacher (parameters are accessible) to train a good student. However, these resources are not always available in real-world app...

Restoration of weather degraded videos is a challenging problem due to diverse weather conditions e.g., rain, haze, snow, etc.. Existing works handle video restoration for each weather using a different custom-designed architecture. This approach has many limitations. First, a custom-designed architecture for each weather condition requires domain-...

Data-free Knowledge Distillation (DFKD) has attracted attention recently thanks to its appealing capability of transferring knowledge from a teacher network to a student network without using training data. The main idea is to use a generator to synthesize data for training the student. As the generator gets updated, the distribution of synthetic d...

Current spacecraft micrometeoroid and orbital debris impact risk assessments utilize semi-empirical equations to describe the protection afforded by a spacecraft component (e.g., pressure hull, critical component, etc.). These equations demand fundamentally limiting assumptions, for example of projectile shape and material, to reduce the complexity...

Objectives: The Future Proofing Study (FPS) was established to examine factors associated with the onset and course of mental health conditions during adolescence. This paper describes the design, methods, and baseline characteristics of the FPS cohort. Methods: The FPS is an Australian school-based prospective cohort study with an embedded cluster...

Knowledge distillation (KD) is an efficient approach to transfer the knowledge from a large "teacher" network to a smaller "student" network. Traditional KD methods require lots of labeled training samples and a white-box teacher (parameters are accessible) to train a good student. However, these resources are not always available in real-world app...

We introduce a novel training procedure for policy gradient methods wherein episodic memory is used to optimize the hyperparameters of reinforcement learning algorithms on-the-fly. Unlike other hyperparameter searches, we formulate hyperparameter scheduling as a standard Markov Decision Process and use episodic memory to store the outcome of used h...

Random Fourier features (RFF) are a popular set of tools for constructing low-dimensional approximations of translation-invariant kernels, allowing kernel methods to be scaled to big data. Apart from their computational advantages, by working in the spectral domain random Fourier features expose the translation invariant kernel as a density functio...

The Kampmann and Wagner numerical model was adapted in MATLAB to predict the precipitation and growth of Al3Sc precipitates as a function of starting concentration and heat-treatment steps. This model was then expanded to predict the strengthening in alloys using calculated average precipitate number density, radius, etc. The calibration of this mo...

The current success of modern visual reasoning systems is arguably attributed to cross-modality attention mechanisms. However, in deliberative reasoning such as in VQA, attention is unconstrained at each step, and thus may serve as a statistical pooling mechanism rather than a semantic operation intended to select information relevant to inference....

We present a Bayesian optimisation methodology intended to support a human expert in the design of armour systems for which limited prior knowledge/data exists and within a limiting, pre-defined experimental budget. We apply the methodology to design an armour configuration consisting of multiple plates, with multiple materials, at varying orientat...

We introduce a conditional compression problem and propose a fast framework for tackling it. The problem is how to quickly compress a pretrained large neural network into optimal smaller networks given target contexts, e.g. a context involving only a subset of classes or a context where only limited compute resource is available. To solve this, we...

Inverse strategies, which incorporate numerical simulation of the experiment and iterative adjustment of model constants until measured signals can be numerically reproduced, are widely employed in finite element modelling, e.g., Finite Element Model Updating (FEMU) [1], Constitutive Equation Gap Method [2], etc. These strategies permit the utilisa...

Solution to many real-world problems often involve the use of expert-level knowledge from various specializations. Such inter-disciplinary problems are usually divided into tasks which are then assigned to a set of bots, each specialized in a particular skill. Supervised selection of the right bot each time is cumbersome and not scalable. Hence the...

Current video enhancement approaches have achieved good performance in specific rainy, hazy, foggy, and snowy weather conditions. However, they currently suffer from two important limitations. First, they can only handle degradation caused by single weather. Second, they use large, complex models with 10-50 millions of parameters needing high compu...

We propose to model the persistent-transient duality in human behavior using a parent-child multi-channel neural network, which features a parent persistent channel that manages the global dynamics and children transient channels that are initiated and terminated on-demand to handle detailed interactive actions. The short-lived transient sessions a...

We introduce a new constrained optimization method for policy gradient reinforcement learning, which uses two trust regions to regulate each policy update. In addition to using the proximity of one single old policy as the first trust region as done by prior works, we propose to form a second trust region through the construction of another virtual...